1052 CORRESPONDENCE
Inorganic Chemistry
Bipyridine Complexes
Sir: We have been determining’ and collecting stability constants of metal complexes with pyridine and its derivatives for many years. The paper oi Atkinson and Bauman2 on the thermodynamics of transition metal complexes with 2,2’-bipyridine therefore received our special attention. We note that even the second decimal for the logarithm of each stability constant is given in this paper. However, there is little agreement between these constants and the values found by other authors3 or obtained in our own l a b ~ r a t o r y . ~ The discrepancies are in some cases higher than one log K unit, which make it obvious that they could not be due to differences in ionic strength. An inspection of the dissertation5 from which the data of ref. 2 have been extracted made i t clear that the authors have underestimated the large errors which arise if a pH method is used in strongly acid solutions. The classical Bjerrum method for determining stability constants is unsuitable for very weakly basic ligands which form metal complexes of considerable stability, such as bipyridine. The findings of Bauman and Atkinson that the degree of complex formation with Mn2+ and Zn2+ rises up to 7t = 5.57 and 6.09, respectively, should have made the authors suspicious of the usefulness of their experimental method. A collection of the more trustworthy stability constants of the bipyridine-transition metal complexes has recently been given by Irving and Mel10r.~ (1) G. Anderegg, Helu. Chim. Acta, 43, 414, 1530 (1960); 46, 1643 (1962). (2) G. Atkinson and J. E. Bauman, Jr., Inoug. Chem., 1 , 900 (1962). (3) H.Irving, private communication. Cited in ref. 2 and Proc. of 1959 Intern. Conf. on Coordination Compounds, London. (4) Unpublished results. For instance: Mnz+, log Ki = 2.5; Cu2+, log K1 = 8.0 a t 20’ and p = 0.1. (5) J. E. Bauman, Thesis, University of Michigan, 1962. (6) H. Irving and D. H. Mellor, J . Chem. Soc., 5222 (1962).
LABORATORIUM FOR ANORGANISCHE CHEMIE EIDG.TECHNISCHEN HOCHSCHULE ZORICH, SWITZERLAND RECEIVED MARCH 27, 1963
G. ANDEREGG
The letter of ,hdet-egg3 refers to unpublished data from his laboratory. Since we have not been so priviliged as to see this data nor been informed as to how i t was obtained, we must reserve comment on it. The Irving-Mellor data referred to by Anderegg have been examined4 and there are some doubts about the validity of the distribution method used by the authors and apparently proposed as a standard by which all other methods and data are to be judged. The method used involved the extraction of free ligand from an aqueous solution containing ligand and metal, buffered with sodium acetate and HC1 and containing KCl to bring i t to what is supposed to be an ionic strength of 0.1. First of all, the authors make no mention of the fact that acetate complexes and chloride complexes of divalent transition metals are well known and, in quite a few cases, their stability constants have been m e a ~ u r e d . ~They also ignore the strong possibility of mixed ligand complexes as well as passing over the real problem of knowing or calculating the ionic strength in such a mixture. Examination of the log K,/k,-l values obtained shows curious inconsistencies not related to obvious problems such as Fe+2spin-pairing and C U +coordination ~ changes. To propose this technique as an interesting and useful approach is certainly valid. To propose i t as a standard by which to judge all others is not justified by their work as presented. Their “recommended values” seem to be only their own data rounded off to two significant figures. (3) G Anderegg, i b i d , 2, 1082 (1963). (4) H. Irving and D. H hlellor, J Chem. Soc., 5222 (1962). ( 3 ) “Stability Constants ” T h e Chemical Society, London, 1957.
DEPARTMENT OF CHEMISTRY USIVERSITY OF MARYLAKD COLLEGE PARK, MARYLAND DEPARTMENT OF CHEMISTRY UNIVERSITY OF MISSOURI COLUXBIA, MISSOURI RECEIVED APRIL 25, 1963
G. ATKINSON J. BAUMAN
Bipyridine Complexes T h e Electronic Spectra of a-Silyl Ketones Sir : Our recent papers1I2 presented a consistent set of A F and AH values for pyridine and bipyridine complexes of Mn+2, Ni+2, C U + ~and , Zn+2. We did not feel then nor do we feel now that this work represents the final word on the thermodynamics of these systems. h’or do we feel that the classical Bjerrum method is totally free of error in log Ki determinations in these systems. We do feel that the thermodynamic data show an internal consistency that is encouraging and that no other method yet proposed is superior to the Bjerrum technique for these systems. (1) G. Atkinson and J. Bauman, Inovg. Chem., 1, 900 (1962). (2) G. Atkinson and J. Bauman, ibid., 2, 64 (1963).
Sir : In 1957 Brook synthesized the first a-silyl ketone, triphenylsilyl phenyl ketone, and established that this compound was yellow in co1or.l This striking observation was followed by an investigation of the electronic spectra of a number of silyl ketones, which showed that in every case the long wave length carbonyl transition was shifted to the visible region.2 Brook attributed the large spectral shift to interaction between lone-pair electrons on the carbonyl oxygen and 3d orbitals of the (1) -4.G. Brook, J . A m . Chem. Soc., 79, 4373 (1957). (2) A. G. Brook, M. A. Quigley, G. J. D. Peddle, E.V. Schwartz, and C. M. Warner, ibid., 82, 5102 (1960).
CORRESPONDENCE 1083
VoZ. 2, No. 5, October, 1963 silicon a t o m 2 We now wish to submit an alternative explanation for the observed shift, based on qualitative molecular orbital theory, which does not iQvolve such ‘‘P a-bonding.” The electronic configuration of the carbonyl group SO and may be simply designated as s ~ ~ awhere ~ p ~ ~ , PO are atomic orbitals localized on the oxygen atom and a is the bonding a-orbital between the carbon and oxygen atoms. The long wave length (n -+ a*) electronic transition in question is considered to result from the transition so2a2p02--t so2a2po&, where a* is the antibonding a-level. Table I lists the absorption maxima of the n + a* transition for several ketonic Compounds with the general formula R(C=O)A. The substituents A may be divided into two classes, A1 and A2. Class 1substituents possess relatively high electronegativities and unshared pairs of electrons and cause shifts to shorter wave length (-Cl, -OH, -NH2); class 2 substituents have low electronegativities and vacant a-type wbitals, and cause shifts to longer yave length ((C&€&Si-, (C&F,)&-, (CeHb)Ge-). It is someiyhat surprising that the shifts caused by class 2 suktituents are about as large as those brought about b y the strongly-interacting type 1 sustituents.
o=c
O=C-A1
Ai
Fig. 1.-Perturbation of energy levels of a carbonyl chromophore by a class 1 substituent, A l .
o=c
O=C-A2
A2
.
,0-
TABLE I LONGWAVELENGTHABSORPTION MAXIMA FOR SOMECARBONYL COMPOUN~ V,.,,
Compound
cm.-l
Ref.
CHaCOOH CHaCONH2 CHaCOCl CHaCOCH3 CHsCaH6 t-CnHeCOCeHs CGHSCOC( CaHs)a CHaCOSi(C6H6)S CaH6COSi( CH3)a CeH6COSi( C6H6)a CaH&OGe( C6H6)s
49,000 46,700 42,600 35,800 30,800 31,000 30,400 26,900 24,200 24,000 24,100
3 3 3 2 2 4 3 2 2 2 2
The influence of substituents of class 1 has been well explained p r e v i ~ u s l y . ~Figure 1 is a simplified, generalized energy level diagram showing the effect of resonant interaction of a class 1 substituent (Al) on the carbonyl transitions. Interaction takes place with both levels T and a*, but most strongly with the bonding level a, because it is closest in energy to TAI. The lowest resultant level a1 is largely localized on A l , so the effect can be viewed as a shifting upward of a and a*. Because POis left essentially unaffected, the net result is to increase the energy of the po + a* transition. Substituents of class 2 should interact by resonance with the carbonyl levels as shown in the generalized diagram of Fig. 2. The 8 8 2 level of the substituent will again interact with both the bonding and antibonding (3) H. H. Jaff6 and M. Orchin. “Theory and Applications of Ultraviolet Spectroscopy,” John Wiley and Sons, Inc., New York, N. Y., 1962, pp. 175-182. Our explanation of the effects of class 1 substituents follows t h a t of Jaffe and Orchin. (4) P. Ramant-Lucas, Bull. SOC.Chim. Fuance, 259 (1950).
Fig. 2.-Perturbation of energy levels of a carbonyl chromophore by a class 2 substituent, A2.
a-levels, but this time most strongly with a*. Here, however, a3* is the orbital localized on the substituent, and the effective result is to shift both a* and a downward in energy. The POlevel is again largely unaffected and the n --t a* transition should fall a t lower energy, as observed.2 Because it is closest in energy to TA2, the a* level is the one most strongly affected by class 2 substituents. Therefore, for a given amount of a-overlap, class 2 substituents will inJtuence the transition energy more than class 1 substituents. Thus the large spectral shift in the silyl ketones can be accounted for, even though the a-3d overlap is probably not large. In the simple argument given above, the purely inductive effect of the substituents has been ignored. The effect of this will be to lower the energies of all of the carbonyl levels for class 1 substituents and to raise them all for class 2 substituents. Unfortunately, i t is difficult to predict which of the two carbonyl levels, a* or PO, will undergo the greatest change. The coulombic interaction of the substituent certainly influences the energy of the n T* transition, but there is reason to believe that these inductive effects are less important than resonance effect^.^^^ Other factors will also influence the energies of carbonyl electronic transitions, including steric effects and, for silyl compounds, perhaps -+
(5) F. A. Matsen, J. Am. Chem. SOC.,72, 5243 (1950).
1084 BOOKREVIEWS
Inorganic Chemistry
even the P-0-Si interaction proposed by Brook and coworkers.2 The latter effect, however, is probably small compared to the C-Si a-bonding considered above. For reasonable values of bond distances and angles in silyl carbonyl compounds, overlap of 3d silicon orbitals is expected to be larger with the carbon a-orbital than with the oxygen lone pairs6 Extension of the approach given above to other chromophoric groups can easily be carried out. For example, class 2 substituents on C=C chromophores should produce a spectral shift of the same type as found for class 1 substituents, decreasing the energy of the a-r* transitions. A further prediction from our theory is that compounds with a boron atom attached (6) Such &interactions may, of course, be important in lowering t h e energies of transition states, L e . , for rearrangement reactions.2~7 (7) A. G. Brook, J . Org. Chem., 25, 1072 (1960); J . Am. Chem. SOC.,80, 1886 (1958); A. G. Brook, C. M. Warner, a n d M. McGriskin, ibid., 81, 981 (1969); A. G . Brook and W. W. Limburg, ibid., 86, 832 (1963).
directly to a carbonyl group (such as R2BCOC6H6), when they are synthesized, will show transitions above 400 mp and so will appear visibly c o 1 0 r e d . ~ ~ ~ Acknowledgments.-The authors thank the Air Force Office of Scientific Research for financial support and the Stauffer Chemical Company for a fellowship (to D. F. H.) (8) Observations on a-boryl ketones should provide a good test of theory, for p r-bonding of t h e type suggested by Brook* should be negligible t o the p-orbital on boron. 19) NOTEADDEDIN PROOF.-A similar, though less detailed, explanation for the spectral properties of rr-silyl ketones has recently been advanced independently by L. E. Orgel, quoted in “Volatile Silicon Compounds,” E. B. Ebsworth, Pergamon Press, Ltd., Oxford, 1963, p. 81.
DEPARTMENT OF CHEMISTRY OF WISCONSIN USIVERSITY MADISOS,WISCONSIN, 53706
DANIELF. HARNISH ROBERT\%‘EST
RECEIVED MAY 31, 1963
Book Reviews Absorption Spectra and Chemical Bonding. By C. K. J@RGENSEN. Addison-Wesley Publishing Co., Inc., Reading, Mass., 1962. 347 pp. 14 X 22 cm. Price, $10.00. Probably the largest amount and certainly the dullest’ prose being written today is the scientific, so-called “literature.” As a notable exception, one might cite the works of Christian Klixbull Jgh-gensen. Former astronomer and sometime symbolic logician, he writes in a style unmistakably his own and often refreshing in its dissimilarity to the usual run of insipid syntax. Of course, roses (excepting Zephirine Drouhin and a few others) come equipped with thorns; that is, one seldom finds unadulterated virtue, with no catches, in this world. Some of Jdrgensen’s prose seems inspired by the stream of consciousness school. Both in his style of writing and in his scientific thinking he is durchaus eigentumlich, with all the good and bad consequences this may have. In the volume under review, his Eigentumlichkeit is much in evidence. I cannot but applaud Jqrgensen’s stated purpose of trying “to win over some of the mental inertia which has caused most chemists to think of chemical bonds in terms of a valency-bond and hybridization description.” I am afraid, though, that the preaching which follows will be mainly t o the already converted, because of the pell-mell attack on the subject matter. The presentation has indeed some logic of its own, but not, I fear, the sort which is appropriate for instructing the uninitiated. Having offered these animadversions, I hasten to say also that this book has its attractions for the serious (but not entirely nai’ve) student of coordination chemistry. It contains a great deal of interesting factual information and many stimulating and unusual observations. It is also noteworthy for its eclectic approach; probably no element in the periodic table goes unmentioned, and all aspects of the subject, as summarized in the title, are discussed to some extent. (1) N o t , however, the most abominable, for which due acknowledgment may be given t o the social “sciences” and “communications sciences.”
MASSACHUSETTS INSTITUTE OF TECHNOLOGYF . A. COTTON CAMBRIDGE 39, MASSACHUSETTS
Progress in Inorganic Chemistry. Volume 4. Edited by F. A. COTTON.Interscience Publishers, Division of John Wiley and Sons, Inc., S e w York, 9.Y., 1962. 540 pp. 15 X 23 cm. Price, $15.00. The fourth volume of the series “Progress in Inorganic Chemistry” contains these seven articles: (1) The Polymorphic Modifications of Arsenic Trioxide, by K. A. Becker, K. Plieth, and I. N. Stanski; (2) The Nephelauxetic Series, by C. K. JIjrgensen ; (3) Peroxides, Superoxides, and Ozonides of the Metals of Groups Ia, IIa, and IIb, by N. Vannerberg; (4) Isopolytungstates, by D. L. Kepert; ( 5 ) Phosphonitrile Polymers, by C. D. Schmulbach; ( 6 ) The trans Effect in Metal Complexes, by F. Basolo and R. G. Pearson; and ( 7 ) The Coupling of Vibrational and Electronic Motions in Degenerate Electronic States of Inorganic Complexes. Part 11. States of Triple Degeneracy and Systems of Lower Symmetry, by A. D. Liehr. This reviewer cannot imagine the inorganic chemist capable of digesting the entire seven-course meal. Articles 1, 3, 4, and 5 are mainly descriptive. They are well written and usefully referenced. Articles 2, 6, and 7 are more theoretical in the order Basolo and Pearson (mildly theoretical), Jdrgensen (highly theoretical), and Liehr (whew!). The account of the trans effect is very readable and complete in every detail. The article on the nephelauxetic series is well worth reading, although it will be rough going for most inorganic chemists. I t is by far the most complete treatment of the subject yet presented. I think it is fair to say that the article overemphasizes many of the author’s own prejudices concerning the electronic structures of transition metal complexes. This is not unexpected since he and C. Schaffer gave us the nephelauxetic fever in the first place. The Liehr article is wasted in this series, being too theoretical for inorganic chemists. It should have been published in one of the volumes read mainly by chemical physicists. Finally, I should mention that a very useful cumulative index (covering volumes 1-4) appears on p. 577. I n summary, “Progress in Inorganic Chemistry” continues to provide excellent critical reviews of interest to inorganic
